linux-kernel - [tip: x86/urgent] x86/entry: Fix noinstr fail

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Tue, 12 Jan 2021 20:12:10 -0000
From:   "tip-bot2 for Peter Zijlstra" <tip-bot2@...utronix.de>
To:     linux-tip-commits@...r.kernel.org
Cc:     Randy Dunlap <rdunlap@...radead.org>,
        "Peter Zijlstra (Intel)" <peterz@...radead.org>,
        Thomas Gleixner <tglx@...utronix.de>, x86@...nel.org,
        linux-kernel@...r.kernel.org
Subject: [tip: x86/urgent] x86/entry: Fix noinstr fail

The following commit has been merged into the x86/urgent branch of tip:

Commit-ID:     9caa7ff509add50959a793b811cc7c9339e281cd
Gitweb:        https://git.kernel.org/tip/9caa7ff509add50959a793b811cc7c9339e281cd
Author:        Peter Zijlstra <peterz@...radead.org>
AuthorDate:    Wed, 06 Jan 2021 15:36:20 +01:00
Committer:     Thomas Gleixner <tglx@...utronix.de>
CommitterDate: Tue, 12 Jan 2021 21:10:58 +01:00

x86/entry: Fix noinstr fail

  vmlinux.o: warning: objtool: __do_fast_syscall_32()+0x47: call to syscall_enter_from_user_mode_work() leaves .noinstr.text section

Fixes: 4facb95b7ada ("x86/entry: Unbreak 32bit fast syscall")
Reported-by: Randy Dunlap <rdunlap@...radead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
Link: https://lore.kernel.org/r/20210106144017.472696632@infradead.org

---
 arch/x86/entry/common.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
index 18d8f17..0904f56 100644
--- a/arch/x86/entry/common.c
+++ b/arch/x86/entry/common.c
@@ -73,10 +73,8 @@ static __always_inline void do_syscall_32_irqs_on(struct pt_regs *regs,
 						  unsigned int nr)
 {
 	if (likely(nr < IA32_NR_syscalls)) {
-		instrumentation_begin();
 		nr = array_index_nospec(nr, IA32_NR_syscalls);
 		regs->ax = ia32_sys_call_table[nr](regs);
-		instrumentation_end();
 	}
 }
 
@@ -91,8 +89,11 @@ __visible noinstr void do_int80_syscall_32(struct pt_regs *regs)
 	 * or may not be necessary, but it matches the old asm behavior.
 	 */
 	nr = (unsigned int)syscall_enter_from_user_mode(regs, nr);
+	instrumentation_begin();
 
 	do_syscall_32_irqs_on(regs, nr);
+
+	instrumentation_end();
 	syscall_exit_to_user_mode(regs);
 }
 
@@ -121,11 +122,12 @@ static noinstr bool __do_fast_syscall_32(struct pt_regs *regs)
 		res = get_user(*(u32 *)&regs->bp,
 		       (u32 __user __force *)(unsigned long)(u32)regs->sp);
 	}
-	instrumentation_end();
 
 	if (res) {
 		/* User code screwed up. */
 		regs->ax = -EFAULT;
+
+		instrumentation_end();
 		syscall_exit_to_user_mode(regs);
 		return false;
 	}
@@ -135,6 +137,8 @@ static noinstr bool __do_fast_syscall_32(struct pt_regs *regs)
 
 	/* Now this is just like a normal syscall. */
 	do_syscall_32_irqs_on(regs, nr);
+
+	instrumentation_end();
 	syscall_exit_to_user_mode(regs);
 	return true;
 }